This project proposes to 1) establish the relation of vascular disease to leukoaraiosis in the deep brain white matter in elderly humans and to correlate leukoaraiosis with heat shock protein (HSP70) accumulation (a new molecular indicator of neurons and glia at risk); and 2) determine the etiology and significance of the ubiquitous exogenous intravascular material (called SCADs) found in brains following cardiopulmonary bypass (CPB). First problem: It is hypothesized that certain abnormal magnetic resonance (MR) signals are associated with intraparenchymal vascular diseases whose ultimate effect is chronic reduction in blood transport capacity. This chronic hypoperfusion may be exacerbated by cervical carotid atherosclerosis. Why important: when these MR abnormalities, often the harbinger of more severe disease, are identified, care should be exercised in these individuals when contemplating therapy or any intervention that might decrease blood or pulse pressure. Second problem: 24-38% of patients will have permanent mild-to-severe intellectual dysfunction following surgery assisted by CPB. Although microemboli appear to be the etiologic agent causing this brain dysfunction, little objective evidence confirming the presence of emboli is reported with the most recent techniques of CPB. Recently, SCADs have been found in autopsy brains shortly after CPB with the alkaline phosphatase (AP) method of vascular staining. SCADs range in size from 10 mu m to 70 mu m, a size that lodges in the smallest vessels of the microvasculature, they have been found in numbers (up to 10-6) that might be expected to cause subtle neurological dysfunction and they are belIeved, but not proven, to be emboli. The outcome of surgery requiring CPB may be improved if SCADs can be eliminated, but first theIr etiology must be establIshed. These cohorts will be studied: l) hypertensives and elderly normotensives; 2) post-CPB patients; and 3) dogs who have undergone CPB. First project: MR of autopsy brain slices will identify areas with leukoaraiosis. Protocol areas will be histochemically stained with the AP technique to facilitate examination of the afferent vasculature at the brain surface for the entire length to and the veins from the lesion to be investigated. Previous pathological correlations of abnormal MR signals deep in the brain have studied only the vessels and neuropil in the immediate abnormal area. It is believed that the proposed study will add significant predictive information to this common MR imaging abnormality. Second project: post CPB autopsy and dog brain material will be studied in order to determine the elemental composition (with laser microprobe mass spectrometry), degree of tissue injury (HSP70 and microinfarcts), and genesis of SCADs.